AI & IP Litigation Tracker
The decisions and laws defining how intellectual property applies to artificial intelligence, organized by the question each one answers, from training-data fair use to AI authorship and digital replicas. Each entry links to our full, source-cited analysis. Updated as new rulings land.
Currently tracking 13 decisions and statutes across 4 fronts. Also see our pillar guide, AI & Intellectual Property.
Last verified: August 3, 2026. Every status in the table below was checked on that date against court records or reputable reporting.
Case status at a glance
How to use this tracker: start with the table for a case's current posture, then follow the link on a case name to our full, source-cited analysis of the underlying decision. The themed sections below the table group those analyses by the question each case answers. Statuses are as of August 3, 2026; in fast-moving dockets, always confirm against the court record before relying on a status.
| Case | Court | Core question | Status (as of August 2026) |
|---|---|---|---|
| Training data and fair use | |||
| Bartz v. Anthropic | N.D. Cal. (filed 2024) | Is training an LLM on copyrighted books fair use, and does using pirated copies change the answer? | Training on lawfully acquired books held fair use (June 2025). The pirated-library claims settled for $1.5 billion, roughly $3,000 per work; final approval granted July 20, 2026, with payouts to follow. |
| Kadrey v. Meta | N.D. Cal. (filed 2023) | Fair use for Llama training, plus liability for downloading books from shadow libraries. | Summary judgment for Meta on training fair use (June 25, 2025) because plaintiffs showed no market harm. Claims over shadow-library acquisition and torrent distribution continue; the court declined to certify an immediate appeal in July 2026. |
| Thomson Reuters v. Ross Intelligence | D. Del. (filed 2020) | Was copying Westlaw headnotes to train a competing legal research AI fair use? | Held not fair use (February 2025). Now the first AI-training fair use case at a federal court of appeals: the Third Circuit heard oral argument June 11, 2026; decision pending. |
| Andersen v. Stability AI | N.D. Cal. (filed 2023) | Do image generators infringe by training on artists’ works and mimicking their styles? | In discovery on the surviving copyright and trademark claims; a third amended complaint was filed February 2026 and a jury trial is set for April 5, 2027. |
| Getty Images v. Stability AI (U.S.) | N.D. Cal. (refiled 2025; originally D. Del. 2023) | Copyright, trademark (watermark), and CMI claims over training Stable Diffusion on Getty’s library. | Getty voluntarily dismissed the Delaware case in August 2025 and refiled in California. On April 23, 2026 the court let the copyright and trademark claims proceed and dismissed only the false-CMI claim. |
| Getty Images v. Stability AI (UK) | High Court, England and Wales (filed 2023) | Does UK copyright reach training done abroad, and are model weights themselves infringing copies? | Decided November 4, 2025. The copyright claims failed (training occurred outside the UK and model weights are not copies); Getty won only narrow trademark findings over generated watermarks. |
| Model outputs, news, and music | |||
| New York Times v. OpenAI (and consolidated cases) | S.D.N.Y. (filed 2023) | Is training on, and regurgitating, paywalled journalism infringement or fair use? | Motions to dismiss largely denied (March 2025); now centralized with other publisher and author suits in MDL No. 3143. Deep in discovery, including a January 5, 2026 ruling upholding production of 20 million de-identified ChatGPT logs. No trial date yet. |
| Concord Music (UMG, ABKCO) v. Anthropic | N.D. Cal. (filed 2023, transferred from M.D. Tenn.) | Song lyrics in training data and in Claude’s output. | Publishers moved for partial summary judgment against Anthropic’s fair use defense in March 2026, and filed a second, far larger suit in January 2026 covering more than 20,000 works. Both cases active. |
| Disney, Universal, and Warner Bros. v. Midjourney | C.D. Cal. (filed 2025) | Are outputs depicting protected characters direct infringement by the image generator itself? | Consolidated with Warner Bros.’ companion suit on November 4, 2025; in active discovery under a schedule that runs expert disclosures through October 2026. |
| RIAA labels v. Suno and v. Udio | D. Mass. / S.D.N.Y. (filed 2024) | Did AI music generators infringe by training on copyrighted sound recordings? | Splitting into deals and fights: UMG and Warner settled with Udio (October and November 2025) and Warner settled with Suno (November 2025), all converting to licenses. Sony continues against both, and UMG expanded its Suno case to 61,026 recordings in May 2026. |
| Dow Jones and NYP Holdings v. Perplexity | S.D.N.Y. (filed 2024) | Does an AI “answer engine” that summarizes articles, sometimes with invented attributions, infringe copyright and trademark rights? | Perplexity’s bid to escape New York jurisdiction failed; the case is in active discovery with a pretrial conference set for December 14, 2026. |
| Ziff Davis v. OpenAI | S.D.N.Y. MDL (filed 2025) | Scraping content despite robots.txt, plus removal of copyright management information. | On December 15, 2025 the court dismissed the DMCA anti-circumvention and unjust-enrichment claims but let the infringement and CMI claims proceed. In July 2026 publishers moved for discovery sanctions against OpenAI. |
| Authorship and inventorship | |||
| Thaler v. Perlmutter | D.C. Circuit | Can an AI system be an “author” under the Copyright Act? | Final. The D.C. Circuit held copyright requires a human author (March 2025), and the Supreme Court denied review on March 2, 2026. |
| Thaler v. Vidal | Federal Circuit | Can an AI system be an “inventor” under the Patent Act? | Final. The Federal Circuit held an inventor must be a natural person (2022); the Supreme Court declined review in 2023. |
| Voice and likeness | |||
| Lehrman v. Lovo | S.D.N.Y. (filed 2024) | Can voice actors sue over unauthorized AI clones of their voices? | Contract and New York right-of-publicity claims survived dismissal (July 2025). Proceedings were stayed on June 11, 2026 after Lovo filed for Chapter 7 bankruptcy. |
AI Training & Fair Use
Whether copying copyrighted works to train AI models is fair use: the central question for every model builder.
Bartz v. Anthropic: Transformative Training, Unforgivable Acquisition Judge Alsup held that training a large language model on books is 'exceedingly transformative' fair use, while refusing to extend that blessing to the pirated library that fed it. The $1.5 billion settlement that followed shows where the real exposure lies. Kadrey v. Meta: A Fair-Use Win That Reads Like a Plaintiffs' Brief Two days after Bartz, Judge Chhabria also found AI training to be fair use, but went out of his way to say the result reflected a failure of advocacy, not a vindication of the practice. His 'market dilution' theory is the doctrine to watch. Thomson Reuters v. Ross: The First Refusal of Fair Use in the AI Era Before the generative-AI rulings, a Delaware court rejected fair use for using copyrighted material to build an AI legal-research tool, and pointedly distinguished the software cases the technology industry had relied upon. Its reach is narrower than its reputation. Andersen v. Stability AI: The Theory That Diffusion Models Can 'Contain' the Works They Trained On A federal court let visual artists' direct and induced copyright claims against AI image generators proceed on the theory that protected works may persist inside the model itself, a holding that reframes how courts think about training data.
AI Output & Infringement
When model outputs reproduce or are substantially similar to training data, and the copyright-management-information claims that travel with them.
The New York Times v. OpenAI: Why a Manhattan Judge Let the Output-Infringement Claims Stand Judge Sidney Stein largely denied OpenAI and Microsoft's motions to dismiss, keeping alive the theory that ChatGPT's outputs and the conduct of its users can infringe. That is a sharp contrast with how California courts have treated similar allegations. Concord Music v. Anthropic: Why the Music Publishers Lost Their Bid to Halt AI Training A federal judge denied the music publishers a preliminary injunction against Anthropic over Claude's reproduction of song lyrics, finding their proposed order overbroad and their claimed harm unproven, even as a guardrails stipulation quietly reshaped the dispute. Stripping the Byline: The § 1202(b) Ruling in The Intercept Media v. OpenAI and the SDNY CMI Split Judge Rakoff let The Intercept's DMCA § 1202(b)(1) copyright-management-information claim against OpenAI survive dismissal while tossing the § 1202(b)(3) claim, splitting from Raw Story v. OpenAI on standing and reshaping CMI litigation in AI-training cases.
AI Authorship & Inventorship
Can an AI be an author or inventor? U.S. authorities have so far required a human.
Authorship Without an Author: Thaler v. Perlmutter and the Human Floor of Copyright The D.C. Circuit held that the Copyright Act requires a human author, foreclosing registration of a work generated autonomously by an AI system. The Supreme Court has now declined to disturb that conclusion. Thaler v. Vidal: Why a Machine Cannot Be a Named Inventor (Yet) The Federal Circuit held that the Patent Act's word 'individual' means a human being, so an AI system called DABUS cannot be listed as an inventor, without deciding whether AI-assisted inventions are patentable at all.
AI Voice, Likeness & Deepfakes
How the right of publicity and new digital-replica statutes apply to AI-generated voices and likenesses.
Lehrman v. Lovo: Why the Right of Publicity, Not Copyright, Governs AI Voice Cloning A federal court let voice actors' right-of-publicity and contract claims against an AI voice-cloning company proceed while dismissing their copyright theories. The decision maps the legal terrain for performers facing synthetic replicas of their voices. Young v. NeoCortext: When a Face-Swap App Meets California's Right of Publicity The Ninth Circuit let a Big Brother contestant's putative class action over the Reface deepfake app proceed, rejecting both a copyright-preemption defense and a transformative-use shield at the pleading stage. The ELVIS Act: Tennessee Makes 'Voice' a Protected Property Right Against AI Cloning Tennessee's 2024 ELVIS Act was the first U.S. law written for AI voice cloning: it made voice an express attribute of Tennessee's right of publicity and reached the tools that clone it. It is a statute, not a court ruling, and a template other states are now copying. California's Digital-Replica Statutes: AB 2602 and AB 1836 Split the Living from the Dead California's two 2024 digital-replica laws take different routes. One voids consent-by-contract for living performers; the other extends the post-mortem right of publicity to AI replicas of the deceased. Together they reset the rules for Hollywood.
Educational content, not legal advice. This tracker summarizes publicly reported
decisions and statutes for general information. AI law is changing rapidly; verify current status
and consult a licensed attorney for your situation.